Sign in Book a demo
Back to Tools
Productivity & Docs OAuth
Build Your Agent View Confluence Docs
confluence logo

Confluence MCP Tools for AI Agents

Let an agent search Confluence, read page content, create pages, and update or rename existing documentation.

Available Tools

15 tools available in this integration

Confluence.CreatePage

Create a new page at the root of the given space.

Confluence.GetAttachmentsForPage

Get attachments for a page by its ID or title.

Confluence.GetAvailableAtlassianClouds

Get available Atlassian Clouds.

Confluence.GetPage

Retrieve a SINGLE page's content by its ID or title.

Confluence.GetPagesById

Get the content of MULTIPLE pages by their ID in a single efficient request.

Confluence.GetSpace

Get the details of a space by its ID or key.

Confluence.GetSpaceHierarchy

Retrieve the full hierarchical structure of a Confluence space as a tree structure

Confluence.ListAttachments

List attachments in a workspace

Confluence.ListDirectChildren

List one page of direct children for a Confluence content item.

Confluence.ListPages

Get the content of multiple pages by their ID

Confluence.ListSpaces

List all spaces sorted by name in ascending order.

Confluence.RenamePage

Rename a page by changing its title.

Confluence.SearchContent

Search for content in Confluence.

Confluence.UpdatePageContent

Update a page's content.

Confluence.WhoAmI

CALL THIS TOOL FIRST to establish user profile context.

Why agent-optimized tools matter

Built for intent, not API schemas

Ask the agent to "find the project runbook, compare it with the onboarding guide, and update the outdated section." Arcade helps it use the right Confluence tools without guessing page IDs, space keys, or hierarchy relationships.

Up to 100x fewer tokens

Confluence search and space-hierarchy reads return focused page, space, and relationship data; fetching several known pages can happen in one request, so the agent need not load an entire Confluence space.

Complete the workflow end to end

Search a space, inspect its hierarchy, read the relevant pages, update the content, and rename the page through one toolkit designed around the complete knowledge-base task.

Make actions predictable and controlled

Arcade tools return structured results and useful errors across users and runs. Teams can apply scope restrictions, redaction, and approval policies so agents follow the same controls every time.

Securing Confluence

The more capabilities agents get, the more useful they are. However, more capabilities bring more risks of undesired effects such as data loss and exfiltration.

This is why Arcade provides you with contextual rules that dictate how and when a tool is allowed to run. Arcade’s contextual access lets you govern tool visibility and behavior at every stage of execution and control who can see a tool, what inputs are allowed, and what comes back.

Learn about Contextual Access Policies

For Confluence, you can implement some of these policies:

Access when tools are listed for a user
  • Hide Confluence.CreatePage from agents outside the roles your IDP marks as content publishers, so unreviewed agents cannot add pages to shared spaces.
  • Deny Confluence.RenamePage to agents that are not operating under your space-owner role, so automation cannot change titles that people use to find approved guidance.
  • Hide Confluence.UpdatePageContent from read-only agents, so they cannot overwrite published procedures or records.
Pre-execution before each tool call
  • Require human approval before Confluence.CreatePage runs in a shared space, so an agent cannot publish an unreviewed page.
  • Block Confluence.RenamePage when the new title removes your required document identifier or approval status, so links and review labels do not become misleading.
  • Refuse Confluence.UpdatePageContent for pages in spaces marked production unless the request comes from your designated content owner, so an agent cannot alter controlled documentation without authorization.
Post-execution after the tool returns
  • Redact credentials, API keys, access tokens, and personal contact details from the page body Confluence.GetPage returns before they reach the model.
  • Strip private download URLs and attachment metadata from Confluence.GetAttachmentsForPage results, so the model cannot expose links to restricted files.
  • Redact page excerpts, comments, and search-result text containing customer records, secrets, or legal-hold material from Confluence.SearchContent before the model sees it.

How to use Confluence with your agent

Pick your MCP client and follow the walkthrough to connect Confluence with per-user OAuth. Then ask your agent to search spaces, inspect page hierarchies, or update documentation.

Your questions, answered

A Confluence MCP server exposes Confluence capabilities as structured tools that an AI agent can call through the Model Context Protocol. After a user authorizes their Atlassian account, the agent can search, read, create, and update permitted Confluence content.

Atlassian provides its own Confluence MCP endpoint for accessing Confluence. Arcade adds user-level OAuth, agent-optimized tools, approval workflows, redaction, and scope controls through one runtime that works across supported clients and frameworks.

Arcade agents can search Confluence, read individual or multiple pages, inspect spaces and page hierarchies, list attachments, retrieve page attachments, create root-level pages, update page content, and rename pages. They cannot delete pages or attachments through this toolkit.

Arcade requests the Confluence OAuth permissions required by the actions you enable. Read-only workflows need read access to the user's permitted Confluence content, while creating, updating, or renaming pages requires write access; the user's Atlassian permissions still limit what the agent can reach.

Yes. Each user authorizes their own Atlassian account through OAuth. Confluence actions run under that user's authorization, so the agent can access only the spaces and pages that user is permitted to access.

Yes. Contextual Access policies can require human approval before consequential Confluence actions such as creating a page, changing page content, or renaming a page. The call proceeds only after the policy's conditions have been satisfied.

Arcade proxies Confluence API calls and does not persist your content as part of normal tool execution. OAuth credentials are brokered and stored encrypted at rest, and the model never receives raw credentials.

Any framework or client that can connect to Arcade's MCP gateway or tool APIs works, including LangChain, OpenAI Agents, CrewAI, Claude Desktop, Claude Code, Cursor, Visual Studio Code, Microsoft Copilot Studio, and GitHub Copilot.

Arcade refreshes Confluence access while the user's authorization remains valid. If access is revoked, rotated, or can no longer be refreshed, actions stop until the user reconnects their Atlassian account.

Atlassian Cloud API rate limits and any limits applied to the user's Confluence organization still apply to calls made through Arcade. Large spaces and page trees may also require multiple paginated reads rather than one unbounded response.

Ready to actually ship your agent?

Connect your agent to Confluence with managed OAuth, agent-optimized tools and approval controls for consequential actions.

Build Your Agent Read the Confluence Quickstart